The anaplastic lymphoma kinase (ALK) is a receptor tyrosine kinase that belongs to the insulin receptor superfamily and is normally expressed in neural tissues during embryogenesis (Morris et al., Oncogene, 1997, 14:2175-2188; Iwahara et al., Oncogene, 1997, 14:439-449). In particular, transcripts of ALK gene are highly expressed in specific regions of the central nervous system, including the diencephalon, midbrain, and the ventral half of the spinal cord. In the peripheral nervous system, ALK expression has been detected in the trigeminal, sympathetic, and enteric ganglia. After birth, expression diminishes, but still persists in certain areas such as the olfactory bulb and thalamus. Despite the apparent function of ALK in the development of the nervous system, the physiologic role of ALK is still largely unclear. While the recent studies are proposing that pleiotrophin (PIN) and midkine (MK) are cognate ligands for ALK (Stoica et al., J Biol Chem, 2001, 276(20):16772-16779; Stoica et al., J Biol Chem, 2002, 277(16):14153-14158), exact mechanisms and biological consequences of ligand-dependent ALK activation are not fully understood at this time.
ALK was initially identified because of its involvement in the human non-Hodgkin lymphoma subtype known as anaplastic large cell lymphoma (ALCL). Many cases of ALCL are associated with a reciprocal translocation, t(2;5)(p23;q35), which juxtaposes the gene at 5q35 encoding nucleophosmin (NPM), a nucleolar-associated phosphoprotein, with the gene for a receptor tyrosine kinase, the anaplastic lymphoma kinase (ALK), at 2p23. The resulting fusion gene encodes a chimeric 80-kD protein in which 40% of the N-terminal portion of NPM is fused to the complete intracytoplasmic portion of ALK containing the functional tyrosine kinase domain (Morris et al., Science, 1994, 263:1281-1284). Constitutive activation of the NPM-ALK kinase domain stimulates anti-apoptotic and mitogenic signaling pathways such as PI3K-AKT, JAK-STAT, and PLCγ, resulting in cellular transformation (Bai, 1998; Slupianek, 2001; Zamo 2002). The transforming activity of NPM/ALK is dependent on its kinase activity (Bischof 1997). While the most frequently occurring oncogenic ALK fusion in ALK-positive ALCL cases (“ALKomas”) is the NPM-ALK (˜80% of ALK-positive ALCL cases), other ALK gene fusions have been consequently identified in human hematological and solid cancers. These include TPM3-ALK (fusion of non-muscle tropomyosin 3 with ALK), TPM4-ALK, ATIC-ALK, CLTC-ALK, RanBP2-ALK, TFGL/S-ALK, CARS-ALK, MSN-ALK and others.
All known ALK fusion proteins share the essential feature of having some type of the oligomerization domain in the sequence of the ALK fusion partner which mediates constituitive self-association of the ALK fusion that causes constant, ligand-independent ALK kinase domain activation. Similarly to NPM-ALK, the related ALK fusion proteins have been shown to possess transforming and oncogenic potential, apparently mediated by their constitutive kinase activity. Although ALK-positive lymphomas have a relatively benign prognosis, about 40% of patients do not respond or relapse after the standard therapy (CHOP). CHOP (cyclophosphamide, hydroxydoxorubicin, oncovin, prednisone) and CHOP-like multi-agent combination chemotherapy regimens that are used for conventional treatment of non-Hodgkin lymphomas including ALCL are associated with considerable acute and chronic toxicities, a problem specifically bothersome in pediatric patients. Therefore, a highly effective and targeted therapy would be beneficial and highly warranted not only for relapsed patients but also as first-line therapy if well tolerated and efficacious.
In addition to ALKomas, several research groups have also described the presence of the NPM-ALK and the related fusion proteins like CLTC-ALK in a rare form of B-cell non-Hodgkin lymphoma. Rearrangements of ALK gene have been also identified in the inflammatory fibroblastic tumors (IMT). These rare spindle cell proliferations involve malignant myofibroblasts and infiltrating non-malignant inflammatory cells in a collagenous matrix and occur primarily in the soft tissue of children and young adults.
More recently, a novel oncogenic ALK fusion, EML4-ALK, comprising portions of the echinoderm microtubule-associated protein-like 4 (EML4) gene and the anaplastic lymphoma kinase (ALK) gene, has been implicated in a subset of non-small cell lung cancer (NSCLC) (Soda, 2007). Mouse 3T3 fibroblast cells forced to express this fusion tyrosine kinase generated transformed foci in culture and subcutaneous tumors in nude mice. The EML4-ALK fusion transcript was detected in 6.7% of the 75 NSCLC patients examined; these individuals were distinct from those harboring mutations in the epidermal growth factor receptor gene. Presence of the oncogenic TPM4-ALK fusion was also detected by proteomics methods in esophageal cancer samples from patients in Iran (Jazii, 2006) and China (Du, 2007). These findings strongly suggest that EML4-ALK and TPM4-ALK fusions are promising candidates for a therapeutic target in a sizable subset of NSCLC and possibly in some esophageal carcinomas.
Certain additional facts concerning the possible relevance of deregulated full-length ALK signaling in some types of cancer and utility of the non-rearranged, full-length ALK as a therapeutic target are noteworthy. The small secreted growth factors pleiotrophin (PTN) and midkine (MK) have been shown to activate signaling of the normal, full-length ALK receptor protein (Stoica et al., 2001, supra; Stoica et al., 2002, supra). While the exact mechanism and biological significance of ALK stimulation by the different molecular forms of these ligands are not completely understood at this time (Lu, 2005; Perez-Pinera, 2007), a functional connection between PTN and/or midkine and ALK is well established. A large number of studies provide evidence that PTN and MK contribute to tumor growth, abnormal tumor-associated angiogenesis and metastasis (Kadamatsu, 2004; Bernard-Pierrot 2002). For example, both PTN and ALK have been found to be overexpressed in human glioblastomas, and downregulation of ALK expression by ribozymes was shown to suppress human glioblastoma xenograft growth in mice and to prolong the survival of the tumor-bearing animals (Powers 2002; Grzhelinsky 2005). Expression or overexpression of the full-length ALK receptor in certain neuroblastomas, diffuse large B-cell non-Hodgkin lymphomas, leiomyosarcomas, and malignant peripheral nerve sheath sarcomas have been reported (Pullford et al., J Cell Physiol, 2004, 199:330-358). Similarly, it has been reported that cell lines established from common solid tumors of ectodermal origin, such as melanoma and breast cancer, exhibit ALK receptor mRNA expression (Pulford, 2004, supra). Additional analyses should elucidate the role of ALK signaling in the genesis and progression of these various cancers over the next few years.
Studies in which the mouse Alk gene was knocked-out demonstrate that ALK-negative mice show no evident gross anatomical, histological or functional abnormalities and have a normal lifespan (Pulford, 2004, supra). Therefore, the physiological functions of Alk, which is normally expressed primarily in neural tissues, appear to be largely redundant. These observations suggest that therapeutic approaches targeting the aberrant oncogenic functions of ALK are not likely to be associated with limiting toxicities due to concomitant inhibition of normal ALK functions.
Therefore, both the various cytoplasmic ALK fusion proteins and the full-length ALK in its transmembrane receptor form are valid molecular targets for anticancer drugs. Consequently, small-molecule inhibitors of ALK kinase are likely to be a drug for suppressing of tumor growth and angiogenesis.
Recently reported preclinical studies have provided compelling proof of principle for the efficacy of the inhibition of NPM-ALK in ALK-positive ALCL, with marked anti-tumor activity observed experimentally. For instance, studies performed by Novartis demonstrated regression of established lymphoma tumors formed by subcutaneous injection of the human NPM-ALK-positive ALCL cell line Karpas-299 in mice when the animals were treated with the small molecule ALK kinase inhibitor NVP-TAE684 (Galkin, 2007).
Other experimental approaches for the inhibition of oncogenic ALK signaling have also indicated that the agents blocking this signaling are likely to possess very potent anti-cancer capabilities. Piva and colleagues recently showed that siRNA (small inhibitory ribonucleic acid)-mediated inhibition of NPM-ALK signaling markedly diminished the development of ALCL xenografts in mice (Piva, 2006). Collectively, these data indicate that the inhibition of the aberrant, cancer-causing activity of ALK fusion proteins in ALCL, as well as other ALK-driven malignancies, using small molecule inhibitors is very likely to produce marked anti-tumor responses.
WO 2004/063151 reported a tyrosine kinase inhibitory activity of pyridones. Pyrroloquinixalinediones and their derivatives were shown to exhibit HIV integrase inhibitory activity (WO2004/096807).
Only a few inhibitors with activity against ALK have been reported. Sauville (Sauville et al, J. Clin. Oncol., 2001, 19, 2319-2333) disclosed a derivative of the natural product staurosporine having an anti-tumor activity in a patient with an ALK-positive anaplastic large cell lymphoma that was refractory to conventional chemo- and radio-therapy. It is important to note that the compound's ability to inhibit ALK was not tested in this study, thus, it has not been formally proven that it is an ALK inhibitor. Indeed, a recent report suggests that staurosporine possesses minimal ability to directly inhibit ALK (Gunby et al., Haematologica, 2005, 90, 988-990). The naturally occurring, structurally related benzoquinone analogues, geldanamycin and 17-allylamino-17-demethoxygeldanamycin (Bonvini et al., Cancer. Res. 2002, 62, 1559-1566) and herbimycin A (Turturro et al., Clin. Cancer Res. 2002, 8, 240-245) have been reported to exert ALK inhibition via heat shock protein pathways, enhancing the proteasome-mediated degradation of the ALK protein. Most recently, a series of pyrazolo[3,4-c]isoquinoline derivatives with ALK-inhibitory activity was published in WO 2005009389.
One of the challenges of developing an ATP-competitive small-molecule ALK inhibitor is to provide sufficient selectivity of the compound for ALK versus inhibition of other structurally related protein kinases. Due to the existence of about 520 evolutionary related protein kinases in the human genome, this could be a demanding task. In particular, inhibition of the insulin receptor kinase which is closely structurally related to ALK is highly undesirable due to the risk of blocking insulin action and the resultant hyperglycemia.
Another highly related RTK is Insulin-Like Growth Factor Receptor I (IGF1R). In the recent years, IGF1R emerged as an attractive oncology target in a broad variety of malignancies (Riedman and Macaulay, 2006; Tao et al 2007). However, suppression of IGF1R signaling may potentially have undesirable side-effects in a clinical context where normal cell/tissue proliferation and development are essential, such as treating pediatric patients (ALCL). Therefore, a sufficiently high selectivity of ALK inhibition versus inhibition of such related RTKs as Insulin Receptor and IGF1R is likely to be a desirable trait in a clinical ALK inhibitor. Conversely, inhibition of a small subset of therapeutically relevant PTKs (multitargeting), in addition to ALK, can improve the efficacy of an oncology drug, especially for solid tumors which are often heterogeneous and have complicated tumor biology.
Another group of tyrosine kinases evolutionary and structurally related to ALK is Ret, Ros, Axl and kinases that are members of Trk family (Trk A, B and C).
RET is a receptor tyrosine kinase that has a role in transducing growth and differentiation signals in tissues derived from the neural crest and is required for normal development of the sympathetic, parasympathetic and enteric nervous systems and the kidney. Gain of function mutations of Ret are associated with the development of several types of human cancers, including medullar thyroid carcinoma and multiple endocrine neoplasias type II and III (or MEN2A and MEN2B). RET mutations have been also identified in a small percentage of pheochromocytomas. Chromosomal rearrangements involving the RET gene are one of the most common causes of a sporadic form of thyroid cancer called papillary thyroid carcinoma (also known as RET/PTC). There is a compelling experimental evidence that thyroid cell transformation to PTC is driven by hyperactivated Ret (Santoro, 2004]. Kinase inhibitors with activity against RET are currently in preclinical or clinical development for these types of cancers.
ROS is a receptor tyrosine kinase that has been found to be constitutively activated in a subset of glioblastomas as a result of genomic translocations (Charest, 2003; Charest, 2006) and may represent an emerging therapeutic target in this highly malignant and deadly brain tumor.
AXL is a unique tyrosine kinase receptor, implicated in the inhibition of apoptosis as well as promoting neovascularization, and it is emerging as a viable therapeutic target in a number of malignancies, both solid and hematologic (Holland, 2005). In particular, it is a chronic myelogenous leukemia-associated oncogene (O'Bryan, 1991; Jannsen, 1991) and is also associated with colon, prostate cancer and melanoma (Van Ginkel, 2004; Sainaghi, 2005). Overexpression of Axl in myeloid cells has been shown to be involved in Type II diabetes (Augustine, 1999). Modulation of Axl activity by small-molecule kinase inhibitors may have utility in therapy of the disease states mentioned above.
TrkA is a receptor tyrosine kinase that belongs to a subfamily of tyrosine kinases that also includes TrkB, and TrkC. TrkB and TrkC are structurally closely related to TrkA, but respond to different ligands in the neurotrophin (NT) family. Nerve growth factor (NGF) signaling through TrkA has been well characterized and is similar to signal transduction mechanisms of other tyrosine kinase receptors. As outlined in more detail below, TrkA is a well validated or a potential drug target in a variety of malignancies as well as in neuropathic pain and certain inflammatory diseases. The roles of the two other members of the neurothropin receptor TK family, TrkB and TrkC, in disease states has received less attention, however the emerging evidence implicates both of them in several types of neoplasias.
TrkA gene was originally described as a chimeric oncogene in colon cancer (Martin-Zanca, 1986) and its activating genomic translocations are common in papillary thyroid carcinomas (Bongarzone, 1989; Pierotti, 2006) and occur in breast cancer as well (Brzezianska, 2007). Hyperactivating deletion or fusion mutations of TrkA and TrkC were also identified in some acute myeloid leukemias as well as solid tumors (Reuther, 2000; Eguchi, 2005).
Overexpression of TrkA in malignant versus normal tissues and association with poor prognosis was shown in prostate, pancreatic cancers, melanomas, mesotheliomas (Festuccia, 2007; Myknyoczki, 1999; Florenes, 2004; Davidson, 2004). TrkA is overexpressed in the majority of prostate carcinomas, and is further increased in androgen-independent tumors (Papatsoris, 2007). In prostatic carcinomas, an autocrine loop involving NGF and TrkA is responsible for tumor progression (Djakiew, 1993). An autocrine NGF/TrkA loop and mitogenic role of NGF has been demonstrated in breast cancer cells as well (Chiarenza, 20011; Dolle, 2003). It has also been shown that NGF signaling has angiogenesis-promoting effect (Cantarella, 2002).
TrkB, sometimes in conjunction with its ligand BDNF, is often overexpressed in a variety of human cancers, ranging from neuroblastomas to pancreatic ductal adenocarcinomas, in which it may allow tumor expansion and contribute to resistance to anti-tumor agents. TrkB acts as a potent suppressor of anoikis (detachment-induced apoptosis), which is associated with the acquisition of an aggressive tumorigenic and metastatic phenotype in vivo (Desmet, 2006; Douma, 2004). In summary, Trk family members have been implicated as oncogenes in a number of neoplasms including prostate, thyroid, pancreatic, colon, breast, ovarian cancers, melanomas and some leukemias. For prostate cancer and thyroid carcinomas, TrkA is especially well validated as a drug target.
Strong and diverse experimental evidence suggests that nerve growth factor (NGF), signaling through TrkA pathway, is a mediator of some persistent pain states, including neuropathic and inflammatory pain (Pent, 2006; Hefti, 2006; Bennet, 2001). Function-neutralizing anti-NGF and anti-TrkA antibodies demonstrated therapeutic effect in models of inflammatory, neuropathic, skeletal and cancer pain (Ugolini, 2007; Koewler, 2007; Sevcik, 2005). In such disease states as prostate cancer with metastatic bone pain and pancreatic cancer with perineural invasion, cancer progression, pain and TrkA signaling has been shown to be all positively correlated (Dang, 2006; Halvorson, 2005). Inhibition of the NGF/TrkA pathway appears to be very well validated for treatment of chronic pain of different natures: (i) inflammatory pain; (ii) neuropathic pain and (iii) cancer pain.
It is noteworthy that in the skin, TrkA receptor mediates the ability of NGF to stimulate keratinocytes proliferation and inhibit keratinocytes apoptosis. NGF is produced by keratinocytes to stimulate their cell proliferation with an autocrine loop and melanocyte proliferation with a paracrine pathway (Di Marco, 1993; Pincelli, 2000). NGF/TrkA signaling also modulates inflammation (Frossard, 2004) and proliferation of terminal cutaneous nerves (Raychaudhury, 2004), components of psoriasis and atopic dermatitis. Murine models for psoriasis and atopic dermatitis have been established and K252a and AG879, both potent non-clinical TrkA inhibitors, were demonstrated to have therapeutic effect [Raychaudhury, 2004) Takano, 2007) in the models. This data indicates that TrkA is a potential drug target in skin disorders characterized by keratinocytes hyperproliferation.
Thus, blocking the ALK activity represents a rational, targeted approach to therapy of various diseases. As there are several tyrosine kinases that are evolutionary and structurally related to ALK, such as Ret, Ros, Axl and members of Trk family, there is an opportunity to either identify a multitargeted kinase inhibitor with a potential utility in other types of malignancies not targeted by selective ALK inhibition, or to fine-tune the inhibition selectivity towards a particular kinase of interest by lead optimization.